GDT HELP
Jazzman56, Thu May 29 2014, 10:40AMHi Guys,
I'm in the middle of building my first ever Solid State Tesla Coil.
• I'm using a modified version of Steve Wards Micro tesla coil.
o Using a 12 Turn GDT like his MINI using a Single UCC37322
o Single IRFP460 to do the power switching
o 75VDC primary (Using a doubler from a 250watt 24ac transformer)
o Using a Feedback Transformer instead of an antenna → cleaned up by a 74hc14 hex inverter.
TESTING suitable Cores
I have tested a variety of cores and picked one, which I though was suitable (out of a microwave oven filter circuit I have no idea what material it is but believe it to be ferrite)
How I tested the Cores:
• I used a oscope probe hooked up to the function generator to the primary side of the GDT to be tested
• I then used another oscope probe into my CRO to look at the resulting waveform.
I Believe it wasn’t the best way to test a GDT as of the impedance of the probes? is this correct ?
Testing Driver board with GDT
I've constructed the driver board and have now run into some problems with the GDT. After switch on the amplitude of the resulting waveform increases for a few seconds, then swings massively, overshooting the limits of my cro's screen. At this point either the UCC37322 or GDT becomes audible and the chip gets hot fast :/ and I think it when poof too :(
• Testing was done with the output of the GDT to a 10ohm (and on another occasion with a 2.7 ohm resistor in parallel) with the probes of my scope across these terminals.
• My function generator can only achieve 200khz although I believe the 555 timer kicks Ucc37322 into oscilation at about 300khz.
Is the Core Reaching Saturation?
How should i go about testing for issues ?
Regards,
Jazzman56
Re: GDT HELP
Thomas W, Thu May 29 2014, 11:19AM
I have had similar things happen to be. This will be because you are using the incorrect core. You want a ferrite one and i do believe this is iron powder.
I cannot, however, remember what core everyone uses more or less...
Thomas W, Thu May 29 2014, 11:19AM
I have had similar things happen to be. This will be because you are using the incorrect core. You want a ferrite one and i do believe this is iron powder.
I cannot, however, remember what core everyone uses more or less...
Re: GDT HELP
Jazzman56, Thu May 29 2014, 11:36AM
Thanks Thomas :P
I'll scrap that core
Does anyone recommend a good type of core material to look for ? Or where i can buy / get one from ? (For both a GDT and Feedback Xformer )
Jazzman56, Thu May 29 2014, 11:36AM
Thanks Thomas :P
I'll scrap that core
Does anyone recommend a good type of core material to look for ? Or where i can buy / get one from ? (For both a GDT and Feedback Xformer )
Re: GDT HELP
Graham Armitage, Thu May 29 2014, 02:16PM
I get mine from EVR -
They work great and are inexpensive.
Graham Armitage, Thu May 29 2014, 02:16PM
I get mine from EVR -
They work great and are inexpensive.
Re: GDT HELP
Hydron, Thu May 29 2014, 07:28PM
Core in the pics looks like ferrite, but the best way to test is to try and use a file on it. Iron powder will be easily cut by the file, ferrite will chip (it's a ceramic) and if you keep going will blunt the file.
Hydron, Thu May 29 2014, 07:28PM
Core in the pics looks like ferrite, but the best way to test is to try and use a file on it. Iron powder will be easily cut by the file, ferrite will chip (it's a ceramic) and if you keep going will blunt the file.
Re: GDT HELP
dexter, Thu May 29 2014, 09:31PM
That core looks like a EMI suppression core which is not suitable for that frequency but people managed to pull it off with even worse cores....
From your description and pictures i see a logic signal (0 - Vcc) fed into the gdt and since there is no DC blocking cap between the ucc and gdt a DC component is always present which is saturating the core and causing all the problems described.
Easy fix pace a 0.1uF or higher polypropylene or ceramic (better) capacitor between ucc and gdt - C6 on Steve's mini SSTC schematic
dexter, Thu May 29 2014, 09:31PM
That core looks like a EMI suppression core which is not suitable for that frequency but people managed to pull it off with even worse cores....
From your description and pictures i see a logic signal (0 - Vcc) fed into the gdt and since there is no DC blocking cap between the ucc and gdt a DC component is always present which is saturating the core and causing all the problems described.
Easy fix pace a 0.1uF or higher polypropylene or ceramic (better) capacitor between ucc and gdt - C6 on Steve's mini SSTC schematic
Re: GDT HELP
Jazzman56, Thu May 29 2014, 11:22PM
Thanks Everyone :P
Unfortunately EVR has a $25 shipping charge. Anyone know where i can get something similar in AUS?
I'll try the DC Blocking cap which i totally forgot to wire in.
Jazzman56, Thu May 29 2014, 11:22PM
Thanks Everyone :P
Unfortunately EVR has a $25 shipping charge. Anyone know where i can get something similar in AUS?
I'll try the DC Blocking cap which i totally forgot to wire in.
Re: GDT HELP
Hydron, Fri May 30 2014, 12:24AM
element14 (formally known as Farnell) is worth looking at, I know they stock some FairRite cores, though not at amazing prices. Other than that, try RS (also likely to be expensive but with low shipping cost), or do a bigger order at Mouser or Digikey to get the free shipping.
As an aside, I feel your pain over shipping costs, NZ is just as bad, and our GST threshold for imports is a lot lower too (400 NZD vs 1000 AUD).
Edit: For the record, I've gotten away with using EMI suppression ferrite toroids for gate drivers (didn't have anything better on hand) - seemed to work fine, but isn't ideal. I just wound 10-12 turns of CAT5 though it and was away laughing. The DC blocking cap is important!
Hydron, Fri May 30 2014, 12:24AM
element14 (formally known as Farnell) is worth looking at, I know they stock some FairRite cores, though not at amazing prices. Other than that, try RS (also likely to be expensive but with low shipping cost), or do a bigger order at Mouser or Digikey to get the free shipping.
As an aside, I feel your pain over shipping costs, NZ is just as bad, and our GST threshold for imports is a lot lower too (400 NZD vs 1000 AUD).
Edit: For the record, I've gotten away with using EMI suppression ferrite toroids for gate drivers (didn't have anything better on hand) - seemed to work fine, but isn't ideal. I just wound 10-12 turns of CAT5 though it and was away laughing. The DC blocking cap is important!
Re: GDT HELP
Sigurthr, Fri May 30 2014, 01:18AM
I believe you have access to Digikey down under, a good friend of mine mentioned ordering from them and he lives in NSW.
I use these cores in all my SSTC work, and they perform flawlessly up beyond 1MHz and down to 30Hz. I've driven large capacitive gates quickly with no issue or sign of saturation.
495-3868-ND
Sigurthr, Fri May 30 2014, 01:18AM
I believe you have access to Digikey down under, a good friend of mine mentioned ordering from them and he lives in NSW.
I use these cores in all my SSTC work, and they perform flawlessly up beyond 1MHz and down to 30Hz. I've driven large capacitive gates quickly with no issue or sign of saturation.
495-3868-ND
Re: GDT HELP
Hydron, Fri May 30 2014, 01:41AM
$30 USD is actually very reasonable for fedex express (fedex can get well into the hundreds shipping small-medium boxes to AUS/NZ), but if you just want to quickly grab a few parts it's very frustrating.
Those cores look good by the way, thanks for the info!
Hydron, Fri May 30 2014, 01:41AM
Sigurthr wrote ...
I believe you have access to Digikey down under, a good friend of mine mentioned ordering from them and he lives in NSW.
Yeah, we can get parts from Digikey/Mouser fairly quickly and easily, it's just not always cost effective for small orders. If we use the US site (in USD) we pay $30 USD for flat rate shipping, and if we use the local NZ/AUS sites the prices are 10-15% higher but with free shipping on orders over ~100USD.I believe you have access to Digikey down under, a good friend of mine mentioned ordering from them and he lives in NSW.
$30 USD is actually very reasonable for fedex express (fedex can get well into the hundreds shipping small-medium boxes to AUS/NZ), but if you just want to quickly grab a few parts it's very frustrating.
Those cores look good by the way, thanks for the info!
Re: GDT HELP
teravolt, Fri May 30 2014, 04:51PM
the whole circuit might be feeding back on its self. add some tantalum to the chips close to the IC power pins. also the correct ferrite core will help. if you are energetic use a proto board with a ground plane. some times those chips have problems driving transformers if it is not done right.
teravolt, Fri May 30 2014, 04:51PM
the whole circuit might be feeding back on its self. add some tantalum to the chips close to the IC power pins. also the correct ferrite core will help. if you are energetic use a proto board with a ground plane. some times those chips have problems driving transformers if it is not done right.
Re: GDT HELP
Ash Small, Fri May 30 2014, 05:09PM
I'd keep looking for a scrap ferrite core. If you can't find a toroid, try a pair of 'E' cores, or something.
Ash Small, Fri May 30 2014, 05:09PM
I'd keep looking for a scrap ferrite core. If you can't find a toroid, try a pair of 'E' cores, or something.
Re: GDT HELP
teravolt, Fri May 30 2014, 09:15PM
something that might also help is to add a parallel 10 ohm and 1uf mono ceramic in series with the transformer.
teravolt, Fri May 30 2014, 09:15PM
something that might also help is to add a parallel 10 ohm and 1uf mono ceramic in series with the transformer.
Re: GDT HELP
Jazzman56, Sun Jun 01 2014, 02:42AM
Hi Guys, I'll try everything above :P
I've added a .1 uf cap too the primary of the GDT helped a bit . The signal is sooo bad my cro can't even trigger correctly :/ to show me the waveform. I believe i'm getting feedback issues. I've added a small 100uf cap to the output of the 5v regulator as it was struggling to supply enough current to the 74hc14.
Ash
I have an ferrite core from a ATX power supply ? Is this suitable ? and if so what is the best way to go about utilising it.
Teravolt
To confirm i'm adding the tantalums to the VCC of the chips then to ground ? Is the the parrallel 10 ohm and series 1uf on the primary or secondary side of the GDT ?
If all else fails i'm making a PCB in eagle :D
Thanks :P
Jazzman56, Sun Jun 01 2014, 02:42AM
Hi Guys, I'll try everything above :P
I've added a .1 uf cap too the primary of the GDT helped a bit . The signal is sooo bad my cro can't even trigger correctly :/ to show me the waveform. I believe i'm getting feedback issues. I've added a small 100uf cap to the output of the 5v regulator as it was struggling to supply enough current to the 74hc14.
Ash
I have an ferrite core from a ATX power supply ? Is this suitable ? and if so what is the best way to go about utilising it.
Teravolt
To confirm i'm adding the tantalums to the VCC of the chips then to ground ? Is the the parrallel 10 ohm and series 1uf on the primary or secondary side of the GDT ?
If all else fails i'm making a PCB in eagle :D
Thanks :P
Re: GDT HELP
Ash Small, Sun Jun 01 2014, 10:04AM
If it's a toroidal one and it's yellow/white or something like that I think it's iron powder, and therefore not suitable.
If it's an 'EE' core, it should be usable. They do say you get better coupling (less losses) with a toroid, but other forms of core should still be ok.
Ash Small, Sun Jun 01 2014, 10:04AM
Jazzman56 wrote ...
Ash
I have an ferrite core from a ATX power supply ? Is this suitable ? and if so what is the best way to go about utilising it.
Ash
I have an ferrite core from a ATX power supply ? Is this suitable ? and if so what is the best way to go about utilising it.
If it's a toroidal one and it's yellow/white or something like that I think it's iron powder, and therefore not suitable.
If it's an 'EE' core, it should be usable. They do say you get better coupling (less losses) with a toroid, but other forms of core should still be ok.
Re: GDT HELP
teravolt, Mon Jun 02 2014, 07:25PM
this is what i ment
teravolt, Mon Jun 02 2014, 07:25PM
this is what i ment
Re: GDT HELP
RateReducer, Sat Oct 11 2014, 01:42AM
Hi. I have a question about cat5 cables and GDTs. I don´t want to open a new thread for only a short question so i hope it is OK to post it here.
Can i also use a shielded cat5 cable e.g. cat5e to wind a GDT or will it be counterproductive?
I got some cat5 cable in an electronic store and i did not notice that it was a cat5e cable... I already wound 2 GDTs with it...
RateReducer, Sat Oct 11 2014, 01:42AM
Hi. I have a question about cat5 cables and GDTs. I don´t want to open a new thread for only a short question so i hope it is OK to post it here.
Can i also use a shielded cat5 cable e.g. cat5e to wind a GDT or will it be counterproductive?
I got some cat5 cable in an electronic store and i did not notice that it was a cat5e cable... I already wound 2 GDTs with it...
Re: GDT HELP
Jazzman56, Fri Oct 24 2014, 09:23AM
Hi Rate Reducer,
I'm pretty sure it is ok to use Cat5 cable to wind GDT's i didn't have any on hand so i use some hookup wire (essentially the same thing but with a larger insulation layer). I'm not sure on shielded wire though.... i though you could get pretty similar results by just twisting individual pairs together.
Regards,
Jazzman56
Jazzman56, Fri Oct 24 2014, 09:23AM
Hi Rate Reducer,
I'm pretty sure it is ok to use Cat5 cable to wind GDT's i didn't have any on hand so i use some hookup wire (essentially the same thing but with a larger insulation layer). I'm not sure on shielded wire though.... i though you could get pretty similar results by just twisting individual pairs together.
Regards,
Jazzman56
Print this page